performance evaluation of multithreaded sparse matrix-vector     multiplication using openmp

	performance evaluation of multithreaded sparse matrix-vector multiplication using openmp
	Liu Shengfei; Zhang Yunquan; Sun Xiangzheng; Qiu RongRong
	2009
会议名称	11th IEEE International Conference on High Performance Computing and Communications
会议录名称	2009 11th IEEE International Conference on High Performance Computing and Communications, HPCC 2009
会议日期	JUN 25-27,
会议地点	Seoul, SOUTH KOREA
出版地	345 E 47TH ST, NEW YORK, NY 10017 USA
出版者	HPCC: 2009 11TH IEEE INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS
ISBN	978-1-4244-4600-1
部门归属	Liu, Shengfei; Zhang, Yunquan; Sun, Xiangzheng Chinese Acad Sci, Inst Software, Beijing 100864, Peoples R China.
摘要	Sparse matrix-vector multiplication is an important computational kernel in scientific applications. However, it performs poorly on modern processors because of a low compute-to-memory ratio and its irregular memory access patterns. This paper discusses the implementations of sparse matrix-vector algorithm using OpenMP to execute iterative methods on the Dawning S4800A1. Two storage formats (CSR and BCSR) for sparse matrices and three scheduling schemes (static, dynamic and guided) provided by the standard OpenMP are evaluated We also compared these three schemes with non-zero scheduling, where each thread is assigned approximately the same number of non-zero elements. Experimental data shows that, the non-zero scheduling can provide the best performance in most cases. The current implementation provides satisfactory scalability for most of matrices. However, we only get a limited speedup for some large matrices that contain millions of non-zero elements.
关键词	Dawning S4800a1 Openmp Compute-to-memory Ratio Irregular Memory Access Patterns Iterative Methods Multithreaded Sparse Matrix-vector Multiplication Nonzero Scheduling Performance Evaluation Application Program Interfaces Matrix Multiplication Multi-threading Scheduling Shared Memory Systems Software Performance Evaluation Sparse Matrices Vectors
主办者	IEEE Comp Soc, IEEE
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/8288
专题	并行软件与计算科学实验室
推荐引用方式 GB/T 7714	Liu Shengfei,Zhang Yunquan,Sun Xiangzheng,et al. performance evaluation of multithreaded sparse matrix-vector multiplication using openmp[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:HPCC: 2009 11TH IEEE INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS,2009.